Al

Fa

Lb

R

a

ARRAA

KTCPdA

1

idnkbwtTs(

sseeasaa

0d

Biosensors and Bioelectronics 26 (2011) 2489–2494

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journa l homepage: www.e lsev ier .com/ locate /b ios

selective novel non-enzyme glucose amperometric biosensor based onectin–sugar binding on thionine modified electrode

eng Lia,b, Yan Fengb, Limin Yangb, Liang Lib, Chenfei Tangb, Bo Tanga,∗

College of Chemistry, Chemical Engineering and Materials Science, Engineering Research Center of Pesticide and Medicine Intermediate Clean Production, Ministry of Education, Keyaboratory of Molecular and Nano Probes, Ministry of Education, Shandong Normal University, 88 Wenhua East Road, Jinan 250014, People’s Republic of ChinaState Key Laboratory Base of Eco-Chemical Engineering, College of Chemistry and Molecular Engineering, Qingdao University of Science and Technology, Qingdao 266042, People’sepublic of China

r t i c l e i n f o

rticle history:eceived 4 August 2010eceived in revised form 8 October 2010ccepted 25 October 2010vailable online 31 October 2010

a b s t r a c t

A novel non-enzyme glucose amperometric biosensor was fabricated based on biospecific binding affin-ity of concanavalin A (Con A) for d-glucose on thionine (TH) modified electrode. TH can be covalentlyimmobilized on potentiostatically activated glassy carbon electrode through Schiff-base reaction. Subse-quently, the surface-adherent polydopamine film formed by self-polymerization of dopamine attached to

eywords:hionineoncanavalin Aolydopamine-Glucose

TH and afforded binding sites for the subsequent immobilization of Con A molecules via Michael additionand/or Schiff-base reaction with high stability. Thus, a sensing platform for specific detection towards d-glucose was established. The binding of Con A towards d-glucose can be monitored through the decreaseof the electrode response of the TH moiety. Due to the high affinity of Con A for d-glucose and high stabil-ity of the resulting sensing platform, the fabricated biosensor exhibited high selectivity, good sensitivity,and wide linear range from 1.0 × 10−6 to 1.0 × 10−4 M with a low detection limit of 7.5 × 10−7 M towards

mperometric biosensor d-glucose.

. Introduction

Diabetes mellitus which is a leading cause of death and disabil-ty nowadays has become a public health concern. Patients withiabetes are suffering from metabolic disorder, also exposed to sig-ificant long-term risks of complications, including heart disease,idney failure, and blindness. According to population studies, dia-etes mellitus has affected approximately 171 million individualsorldwide in the year 2000 and it will reach up to 366 million by

he year 2030 with an exponential prevalence (Wild et al., 2004).herefore, the determination of blood glucose level is of great ofignificance for the diagnosis and management of diabetes mellitusPickup et al., 2005).

Numerous glucose biosensors were developed for glucose mea-urement since the first glucose biosensor was developed. Turbidityensors (Ballerstadt et al., 2007), viscometric affinity sensors (Diemt al., 2004), surface plasmon resonance based sensors (Gallegot al., 2004), fluorescence based sensors (Russell and Pishko, 1999),

nd electrochemical biosensors (Li et al., 2009a) have been exten-ively studied. Among them, the electrochemical biosensor can begood choice for its simplicity, rapid response, high sensitivity,

nd low cost. As a member of electrochemical biosensors, enzy-

∗ Corresponding author. Fax: +86 531 86180017.E-mail address: tangb@sdnu.edu.cn (B. Tang).

956-5663/$ – see front matter © 2010 Published by Elsevier B.V.oi:10.1016/j.bios.2010.10.040

© 2010 Published by Elsevier B.V.

matic glucose electrochemical biosensors based on modification ofglucose oxidase on various substrates are of particular importancedue to the biospecific affinity of enzyme for its substrate. However,friendly environment is essentially required to retain the activityof the enzyme (Li et al., 2009b), leading to a higher requirement onmaterial selection and operation conditions. Furthermore, there issevere interference caused by endogenous electroactive uric acid(UA) and ascorbic acid (AA) in blood samples (Safavi et al., 2008).The two points above limit the application of enzymatic biosen-sors in a way. As a result, non-enzyme based glucose biosensorshave been receiving increased attention in recent years (Aslan et al.,2004; Ma et al., 2009).

Usually, non-enzyme based glucose biosensors are fabricatedbased on the specific affinity of certain species for glucose, includingboronic acid derivatives (Yu and Yam, 2009), antibodies (Engstromet al., 2008), and lectins (Nakata et al., 2005). Concanavalin A (ConA), one member of lectins family, has been extensively used inglucose recognition. It appears to be dimeric below pH 5.5, andundergoes the dimer–tetramer equilibrium between 5.5 and 7.5. AtpH > 7.5, the protein exists as a tetramer (molecular mass, 104,000for tetramer) (Seneart and Teller, 1981). Under neutral conditions,

each subunit of Con A contains one binding site for d-mannoseand d-glucose residues; one for calcium and manganese cations,which activate the binding site of protein for carbohydrates; thethird one for hydrophobic recognition (Liu et al., 2007). Glucosemeasurements based on Con A–glucose binding were commonly

2 oelect

csntbtob

sa(damanisnodltmefSa

mCiaTwbdbatta

2

2

pcbuvNHd

2

(cEC

490 F. Li et al. / Biosensors and Bi

onducted by fluorescence biosensors, due to its intrinsic high sen-itivity, however, the sensor performance was often influenced byon-specific changes in fluorescence (Labib et al., 2010). Alterna-ively, it will be an efficient and reliable way to apply Con A–glucoseinding to the fabrication of glucose amperometric biosensor. Tohe best of our knowledge, there is no publication on the usagef Con A–glucose binding for fabrication glucose amperometriciosensor yet.

The efficient immobilization of Con A on underlying sub-trates is always a challenge, as Con A will denature and loosectivity when it is adsorbed on bare surfaces. Dopamine3,4-dihydroxyphenylethylamine) is an analogue of 3,4-ihydroxyphenylalanine which plays a major role in strongdhesion of a mussel onto a variety of substrates in tidal environ-ent (Lee et al., 2008a). In the presence of dissolved O2 and under

lkaline conditions, aqueous solutions of dopamine will sponta-eously oxidize in about 10 min to polydopamine (pDA) which

s an adherent polymer coating on both organic and inorganicurfaces (Lee et al., 2008b). The DA self-polymerization mecha-ism was not clear until now. It is acceptable that DA is firstlyxidized to dopaminequinone; the intramolecular cyclization ofopaminequinone via 1,4-Michael addition leads to oxidizable

eucodopaminechrome; then leucodopaminechrome is oxidizedo dopaminechrome. Dopaminechrome can further undergo poly-

erization reactions, yielding deposited melanin-like polymer (Lit al., 2006). The resulting pDA film can be an appealing materialor immobilization of Con A through Michael addition and/orchiff-base reaction between the quinone/catechol groups of pDAnd the ε-NH2 of the lysine residues of Con A (Morris et al., 2009).

Herein, a novel selective non-enzyme based glucose ampero-etric biosensor was fabricated based on the specific affinity of

on A for d-glucose. TH acted as the electrochemical responsendicator, which was covalently immobilized on potentiostaticallyctivated glassy carbon electrode (GCE). Then, pDA was formed onH modified surface as an adhesive coating in alkaline solution,hich also afforded binding sites for the adhesion of Con A. The

inding between Con A and d-glucose was monitored through theecrease of the electrochemical response of the TH moiety. The sta-ility of the sensing platform, the high affinity of Con A ford-glucoses well as the excellent electron transfer ability of TH endowedhe as-prepared glucose biosensor with low detection limit, rela-ively wide linear range, good reproducibility, long-term stabilitynd high anti-interference ability.

. Experimental

.1. Chemicals

Dopamine hydrochloride, dextran (39 kDa) and d-glucose wereurchased from Sigma–Aldrich and used as received. TH was pur-hased from Guo Yao Company (China). Con A (Shanghai sanjieiotechnology Co. Ltd.) from canavalin ensiformis (Jack Bean) wassed in this study. The 0.1 M phosphate buffer solutions (PBS) atarious pH values were prepared by mixing the stock solutions ofaH2PO4 and Na2HPO4, and adjusting the pH with 0.1 M NaOH or3PO4. All other chemicals were of analytical grade and doubly-istilled water (DDW) was used throughout this work.

.2. Apparatus and instruments

Cyclic voltammetric (CV) and differential pulse voltammetricDPV) measurements were performed on a CHI 832B electro-hemical analyzer (Shanghai CH Instrument Company, China).lectrochemical impedance spectroscopy (EIS) was performed on aHI 660C electrochemical analyzer (Shanghai CH Instrument Com-

ronics 26 (2011) 2489–2494

pany, China). A conventional three-electrode system was used withbare GCE or modified GCE as the working electrode, Ag/AgCl elec-trode (saturated with KCl) as the reference electrode, and platinumwire as the auxiliary electrode, respectively.

2.3. Preparation of pDA film on TH modified GCE

GCE was used as the substrate electrode in the construction ofthe biosensor. Prior to use, the bare GCE was polished with 1.0, 0.3,and 0.05 �m alumina slurry to obtain a mirror like surface, respec-tively, followed by thorough rinsing with DDW, and sonicating inDDW and ethanol. Potentiostatic activation was carried out by oxi-dizing the polished electrodes in 0.1 M HNO3 at 1.8 V for 3 min, andthen reducing at −1.5 V for 1 min (Chen et al., 2009). For covalentimmobilization of TH, the obtained electrode above was immersedin 2 mg/mL TH solution (0.1 M PBS pH 7.0) for 3 h, and then washedby ultrasonication for 10 min, followed by rinsing with 0.1 M PBS(pH 7.0) to remove non-specifically adsorbed TH. The as-preparedelectrode was denoted as TH/GCE.

Subsequently, the TH/GCE was immersed into 6 mM dopaminesolution (0.1 M PBS pH 8.5) for 2 h under stirring (Lee et al., 2007).Then, the electrode was thoroughly rinsed with DDW and driedwith nitrogen, and denoted as pDA/TH/GCE.

2.4. Immobilization of Con A on pDA modified surface and thebinding of d-glucose

10 mg Con A was dissolved in 10 mL 0.1 M PBS (pH 7.0) con-taining 0.1 M CaCl2 and 0.1 M MnCl2 and kept for 12 h at 4 ◦C toobtain 1 mg/mL Con A solution. The immobilization of Con A wasperformed by incubating pDA modified electrode in 1 mg/mL ConA solution under stirring for 2 h at room temperature (Morris et al.,2009). After the immobilization step, the electrode was carefullyrinsed with 0.1 M PBS (pH 7.0) and Con A/pDA/TH/GCE was thusobtained. To examine the interaction between d-glucose and ConA, the Con A/pDA/TH/GCE was incubated in 0.1 M PBS (pH 7.0)containing various concentrations of d-glucose under stirring. Theincubation time of Con A binding for d-glucose was also investi-gated.

2.5. Characterization of the developed biosensor

The modified electrodes obtained during stepwise modificationwere electrochemically characterized by using CV, EIS and DPV. CVand DPV measurements were performed in 0.1 M PBS (pH 7.0) qui-escent solutions with a scan rate of 50 mV s−1. EIS measurementwas performed with the frequency ranging from 104 to 10−1 Hzin 0.1 M KCl containing 5.0 mM K3Fe(CN)6/K4Fe(CN)6 (1:1) withthe AC voltage amplitude was 5 mV, and the applied potential of221 mV versus Ag/AgCl electrode.

3. Results and discussion

3.1. Immobilization of Con A on pDA modified electrode and thebinding of d-glucose

As is known to all, Con A has specific affinity for d-glucose andd-mannose, based on which a novel glucose amperometric biosen-sor was constructed for selective recognition of d-glucose. As isillustrated in Fig. 1, TH was firstly immobilized on CHO-richedsurface which was generated via a potentiostatic activation pro-

cess in diluted nitric acid. Potentiostatically activated GCE surfacepossessed microporous structure (Nagaoka and Yoshino, 1986),leading to an increased electrode area available for loading of TH.The in situ generated-COH reacted with TH to form imine bond,which contributed to the stable attachment of TH on electrode

F. Li et al. / Biosensors and Bioelectronics 26 (2011) 2489–2494 2491

abrica

stitcopIaot�aaepsaaic

3

(Ps(a(eswip

−0.6 V, and the current obtained on the as-prepared GCE suffersfrom almost no degradation. It is probable that pDA has stabi-lization effect on the sensing surface, which efficiently avoids thedissociation of TH from the electrode surface.

Fig. 1. Schematic presentation of the f

urface. Then, pDA adhesion layer was formed on the amine-erminated electrode surface just by immersion of the electroden dopamine alkaline solution (Lee et al., 2007). It can be seenhat the resulting electrode surface was covered by a brown-olored melanin-like polymer, indicating the successful formationf pDA film. Afterwards, Con A was covalently adhered to theDA film through Michael addition and/or Schiff-base reactions.

n fact, the nature of the interaction between the bound Con And pDA is not quite clear. It was acceptable that the N atomf ε-NH2 on the lysine residues of Con A owns lone pair elec-rons, thus –NH2 can nucleophilically attack the �-carbon of an,�-unsaturated quinone moiety of the pDA polymer, yieldingmine–quinone adducts. Alternatively, –NH2 can nucleophilicallyttack the C atom of the unsaturated carbonyl on the quinine moi-ty, leading to carbon–nitrogen double bonds between Con A andDA (Morris et al., 2009; Burzio and Waite, 2000). The pDA adhe-ive coating had the stability similar to chemical bond coating (Yinnd Liu, 2008), which can dramatically improve stability of TH layers well as Con A on electrode surface, leading to enhanced stabil-ty of the sensing surface. The resulting sensing platform was thusonstructed for specific determination of d-glucose.

.2. Electrochemical behaviors of the glucose biosensor

Fig. 2 shows CVs of (a) bare GCE, (b) TH/GCE, (c) pDA/TH/GCE,d) Con A/pDA/TH/GCE, and (e) glucose/Con A/pDA/TH/GCE in 0.1 MBS (pH 7.0) over the potential range from 0.1 to −0.6 V at acan rate of 50 mV s−1. No redox peak appeared on the bare GCEFig. 2a), while a pair of well-defined redox peaks was observedt TH/GCE (Fig. 2b), with the anodic (Epa) and cathodic potentialEpc) of −0.22 V and −0.26 V, respectively. This can be undoubt-

dly attributed to the redox reaction of TH, indicating TH has beenuccessfully immobilized on electrode surface. After pDA layeras formed on the TH/GCE via a spontaneous oxidization coat-

ng technique, the signal of the peak currents decreased, and theeak-to-peak separation (�Ep) was increased (Fig. 2c), which was

ted glucose amperometric biosensor.

probably that the pDA film had obstructed the electron transfer.Another oxidation peak located at ca. −0.04 V can be ascribed tothe oxidation of leucodopaminechrome. When the Con A moleculeswere immobilized onto the pDA/TH/GCE (Fig. 2d) and the sub-sequent binding for d-glucose (Fig. 2e), the peak current furtherdecreased, implying the attachment of Con A and glucose hasenhanced the inhibition effect on the electron transfer. It has tobe mentioned that the modified GCE in neutral solutions is quitestable during repetitive CV scans over the potential range of 0.1 to

Fig. 2. CVs (a) bare GCE, (b) Th/GCE, (c) pDA/Th/GCE, (d) Con A/pDA/Th/GCE, and(e) glucose/Con A/pDA/Th/GCE in 0.1 M PBS (pH 7.0) at a scan rate of 50 mV s−1;concentration of glucose: 4.0 × 10−5 M.

2492 F. Li et al. / Biosensors and Bioelectronics 26 (2011) 2489–2494

FAF

rPAWttAa(

ti(oFnooiotFi3ttfiglr

3

co0tstwt

ig. 3. EIS for (a) bare GC electrode, (b) Th/GCE, (c) pDA/Th/GCE, (d) Con/pDA/Th/GCE, and (e) glucose/Con A/pDA/Th/GCE in 0.1 M KCl containing 5.0 mMe(CN)6

3− and 5.0 mM Fe(CN)64− .

Further study was performed to investigate the effect of scanate on the response of TH at the Con A/pDA/TH/GCE in 0.1 MBS (pH 7.0) (see supplementary material Fig. S1). CVs of Con/pDA/TH/GCE gave nearly symmetric anodic and cathodic peaks.ith the increasing scan rate, Epa and Epc were shifted slightly

owards the positive and the negative direction of potential, respec-ively. Both the anodic and the cathodic peak currents of the Con/pDA/TH/GCE were proportional to the scan rate between 20nd 500 mV s−1, indicating a surface-controlled electrode processGosser, 1993).

EIS was employed to characterize the interface properties ofhe modified electrodes. Fig. 3 shows the typical results of ACmpedance spectra of the bare GCE (a), TH/GCE (b), pDA/TH/GCEc), Con A/pDA/TH/GCE (d), and glucose/Con A/pDA/TH/GCE (e)btained in 0.1 M KCl containing 5.0 mM Fe(CN)6

3− and 5.0 mMe(CN)6

4− with the frequency ranging from 104 to 10−1 Hz. Sig-ificant differences in the electron transfer resistance (Ret) werebserved upon the stepwise modification of the electrode. Ret valvef the bare GCE (Fig. 3a) was estimated to be 504 �. After covalentmmobilization of TH onto the surface of pretreated GCE, the valuef Ret was decreased noticeably (Fig. 3b), which is attributed tohe strong electrostatic adsorption between the negatively chargede(CN)6

3−/4− and the positively charged TH molecules. The coat-ng of pDA film on the surface of TH/GCE induced the Ret to be261 � (Fig. 3c), implying that the pDA film could largely restricthe electron transfer from the electrochemical probe towards elec-rode surface. The immobilization of Con A molecules onto the pDAlm also led to an increased Ret valve (Fig. 3d), and the binding oflucose made the Ret valve even larger (Fig. 3e), indicating botharge molecule Con A and the binding of glucose can obstruct theedox probe approaching towards the electrode surface.

.3. Influence of pH on the biosensor responses

TH is a pH-dependent dye (Nicotra et al., 2008), so the electro-hemical behavior of TH can be significantly influenced by the pHf supporting solutions. Fig. 4 shows CVs of Con A/pDA/TH/GCE in.1 M PBS from pH 5.0 to 9.0 at a scan rate of 50 mV s−1. It is wor-

hy to note that stable CV curves were obtained during successivecanning cycles in supporting solutions of different pH, suggestinghat the proposed biosensor can be used over a wide pH range. Itas found that both Epa and Epc shifted to a negative direction with

he increase of pH, indicating a both proton and electron transfer

Fig. 4. CVs of the Con A/pDA/Th/GCE in 0.1 M PBS at various pHs: (a) 5.0, (b) 6.0, (c)7.0, (d) 8.0, and (e) 9.0 at a scan rate of 50 mV s−1.

process (Xu et al., 1998). The �Ep changed slightly, indicating agood reversibility of TH on the well-confined surface over the widepH range. It can be seen that the maximum response of TH at theCon A/pDA/TH/GCE was obtained at pH 7.0. Taking account of thepeak current response, the activity of the protein and condition ofspecific binding affinity of Con A for glucose (Sugawara et al., 2004),pH 7.0 was selected for further experiments.

3.4. Determination of d-glucose

Herein, a Con A/TH modified GCE was constructed for specificdetermination of d-glucose. When Con A modified electrode wasimmersed in d-glucose solution, the TH moiety was held at thebinding site of Con A (Sugawara et al., 2004). The introduction ofglucose into the binding site of Con A makes it difficult for theelectron transfer between the redox TH and the electrode surface,which can be reflected by the decreased amperometric response ofTH.

The incubation time of Con A modified electrode in d-glucosesolution can influence the electrochemical response of the biosen-sor. In the experiment, different current signals of TH were obtainedby immersion of the Con A/pDA/TH/GCE in 4.0 × 10−5 M d-glucosefor 20, 40, 60, 80, and 100 min, respectively (data not shown forbrief). With prolonging the incubation time to 60 min, the peak cur-rent change versus that of Con A/pDA/TH/GCE increased, and thepeak current valve trended to almost a constant from 60 to 100 min.Thus, an incubation time of 60 min for Con A modified electrodeimmersion in d-glucose solution was chosen for experiment.

Fig. 5 shows DPV responses of Con A/pDA/TH/GCE to differ-ent concentrations of d-glucose by immersion of the modifiedelectrode in d-glucose solution (0.1 M PBS pH 7.0) for 60 minbefore measurement. It can be found that each oxidationwave for the TH moiety was a well-defined peak and theoxidation current decreased continuously as a function of glu-cose concentration in the range of 1.0 × 10−6 to 1.0 × 10−4 M(shown in inset). The regression equation was expressed as �Ipa

(10−6 A) = −0.04279 C (10−6 M) − 0.42308 (n = 8), with a correlationcoefficient of −0.9969. A detection limit of 7.5 × 10−7 M d-glucosecan be estimated using 3�.

In order to demonstrate the analytical capacity of the pro-posed biosensor, a comparison of linear range and low detectionlimit with other non-enzyme based glucose sensors was listed inTable 1. Compared with some electrochemical sensors based onoxidation of glucose on metal or metal oxide modified electrodes

F. Li et al. / Biosensors and Bioelectronics 26 (2011) 2489–2494 2493

Table 1Comparison of analytical performance of non-enzyme based glucose sensors.

Sensor Linear range (�M) Detection limit (�M) References

Cu/CNTs/GCE 0.7–3500 0.2 Kang et al. (2007)

(Algps2atpws

3

bttfios4iaoaacce

Fg((a

Cu2O/MWCNT 0.05–5.0Nanogold–plasmon-resonance–Con A 1000–40000Fluorescein-labeled Con A–glycogen conjugate 5000–50000Con A/pDA/TH/GCE 1.0–100

Kang et al., 2007; Zhang et al., 2009), the biosensor based on Con-glucose binding exhibited comparable performance, and it estab-

ished a new route for glucose detection without direct oxidation oflucose-self. Compared with the sensors based on the same princi-le of Con A-glucose binding but with different detection methods,uch as surface plasmon resonance and fluorescence (Aslan et al.,004; Sato and Anzai, 2006), the proposed biosensor definitely hasrelatively poor performance, however, it was an attempt by using

he Con A-glucose binding for glucose detection via amperometricrotocol. Although the proposed biosensor possessed a linear rangehich is not in the realm of blood sugar level, it can be used in real

ample analysis just by diluting the sample.

.5. Interferences

One of the most important analytical factors for a glucoseiosensor is the ability to discriminate the interfering species fromhe target analyte. AA and UA presented in physiological fluids arehe most important interferences for analysis of glucose, especiallyor non-enzymatic biosensors (Yuan et al., 2005). Herein, we stud-ed the interference effect of 1.0 × 10−4 M UA and 1.0 × 10−4 M AAn the response of 4.0 × 10−5 M glucose. Experimental resultshowed that AA and UA did not give any observed interference to.0 × 10−5 M glucose of the biosensor (data not shown for brief),

ndicating that the biosensor possesses acceptable anti-interferentbility and excellent selectivity. As the principle for determinationf d-glucose is based on the lectin–sugar binding, so it is reason-

ble to obtain the above results. It has to be mentioned that Con Also has different binding affinities for other sugar molecules, espe-ially d-mannose, d-maltose, methyl-�-glucopyranoside, whichan interfere with glucose detection. However, the interferenceffect on the measurement for glucose can be negligible due to their

ig. 5. DPV responses of Con A/pDA/Th/GCE to different concentrations of d-lucose: (a) 0, (b) 1.0 × 10−6 M, (c) 1.0 × 10−5 M, (d) 2.5 × 10−5 M, (e) 4.0 × 10−5 M,f) 5.5 × 10−5 M, (g) 7.0 × 10−5 M, (h) 8.5 × 10−5 M, and (i) 1.0 × 10−4 M in 0.1 M PBSpH 7.0). Inset shows the linear relationship between the difference of peak currentnd the concentration of d-glucose.

0.1 Zhang et al. (2009)– Aslan et al. (2004)– Sato and Anzai (2006)0.75 Proposed

low concentrations under physiological conditions (Labib et al.,2010).

3.6. Reproducibility and stability of the biosensor

The reproducibility and stability are two important parametersfor the evaluation of the performance of biosensors. The repeata-bility and stability of the proposed biosensor were examined bymeasuring the current response to 4.0 × 10−6 M d-glucose. Theproposed biosensor showed an acceptable repeatability with a rel-ative standard deviation (R.S.D.) of 3.1% for 11 successive assays byregeneration the Con A/TH modified GCE with 1.0 × 10−3 M dex-tran (39 kDa) after the biosensor binds of d-glucose. Six biosensorsprepared independently using the same procedure gave a R.S.D. of3.8%, indicating the biosensor has good reducibility. The stability ofthe Con A/pDA/TH/GCE was evaluated by intermittently measure-ment of 4.0 × 10−6 M d-glucose every 3 days, and it retained 91%of its initial response after 2 months storage. It is probable that thecovalent immobilization of TH preserved TH suffering from loss.The exterior pDA film also has stabilization effect on TH layer andCon A, which dramatically enhanced the stability of the sensingplatform.

3.7. Glucose determination in serum samples

This glucose biosensor could be applied to assay the real bloodsugar. Three fresh serum samples were supplied by the AffiliatedHospital of Qingdao University and the samples were 200-folddiluted with 0.1 M PBS at pH 7.0 before analysis with the biosen-sor. By using standard addition method, 4.0 × 10−5 Md-glucose wasadded into the diluted samples, and measured current was con-verted to the glucose concentration based on the working curve.The recovery for the determination of glucose was in the rangeof 97.8–104.3% for three serum samples, indicating the proposedbiosensor can be applied in real sample determination.

4. Conclusions

Selective determination of d-glucose was realized by a non-enzyme glucose amperometric biosensor based on Con A–glucosebinding with TH as the response indicator. Due to the specificadherence mechanism of pDA through Michael addition and/orSchiff-base reaction with the amino group of TH and Con A,the sensing platform exhibited high stability, which was demon-strated by the fact that the signal of TH hardly decreased duringmulti-scannings and the good long-term stability of the biosen-sor. The specific lectin–sugar binding also endowed the biosensorwith excellent anti-interference ability, and the biosensor showedacceptable recoveries in real serum sample determination. Thenon-enzyme glucose amperometric biosensor holds great potentialin clinic blood sugar detection.

Acknowledgements

This work was supported by the National Basic Research Pro-gram of China (973 Program, 2007CB936000), the National Natural

2 oelect

SM((PtZ(

A

t

R

AB

BC

D

E

G

G

494 F. Li et al. / Biosensors and Bi

cience Funds for Distinguished Young Scholar (No. 20725518), theajor Program of National Natural Science Foundation of China

No. 90713019), the National Natural Science Foundation of ChinaNos. 20875058, 20775039), Science and Technology Developmentrograms of Shandong Province of China (No. 2008GG30003012),he Natural Science Foundation of Shandong Province of China (No.R2009BM031), and the Science Foundation of China PostdoctorNo. 20100471568).

ppendix A. Supplementary data

Supplementary data associated with this article can be found, inhe online version, at doi:10.1016/j.bios.2010.10.040.

eferences

slan, K., Lakowicz, J.R., Geddesa, C.D., 2004. Anal. Biochem. 330, 145–155.allerstadt, R., Kholodnykh, A., Evans, C., Boretsky, A., Motamedi, M., Gowda, A.,

McNichols, R., 2007. Anal. Chem. 79, 6965–6974.urzio, L.A., Waite, J.H., 2000. Biochemistry 39, 11147–11153.hen, X.H., Hua, J.Q., Chen, Z.W., Feng, X.M., Li, A.Q., 2009. Biosens. Bioelectron. 24,

3448–3454.iem, P., Kalt, L., Haueter, U., Krinelke, L., Fajfr, R., Reihl, B., Beyer, U., 2004. Diab.

Technol. Ther. 6, 790–799.

ngstrom, H.A., Johansson, R., Koch-Schmidt, R., Gregorius, K., Ohlson, S., Bergstrom,

M., 2008. Biomed. Chromatogr. 22, 272–277.allego, R.G., Haseley, S.R., van Miegem, V.F.L., Vliegenthart, J.F.G., Kamerling, J.P.,

2004. Glycobiology 14, 373–386.osser Jr., 1993. Cyclic Voltammetry, Simulation and Analysis of Reaction Mecha-

nisms. VCH, Weinheim.

ronics 26 (2011) 2489–2494

Kang, X., Mai, Z., Zou, X., Cai, P., Mo, J., 2007. Anal. Biochem. 363, 143–150.Labib, M., Hedström, M., Amin, M., Mattiasson, B., 2010. Anal. Chim. Acta 659,

194–200.Lee, H., Dellatore, S.M., Miller, W.M., Messersmith, P.B., 2007. Science 318, 426–430.Lee, H., Lee, Y., Statz, A.R., Rho, J., Park, T.G., Messersmith, P.B., 2008a. Adv. Mater.

20, 1619–1623.Lee, S.R., Sawadaa, K., Takao, H., Ishida, M., 2008b. Biosens. Bioelectron. 24, 650–656.Li, F., Wang, Z., Chen, W., Zhang, S.S., 2009a. Biosens. Bioelectron. 24, 3030–3035.Li, F., Wang, Z., Feng, Y., 2009b. Sci. China Ser. B: Chem. 52, 2269–2274.Liu, S.Q., Wang, K.W., Du, D., Sun, Y.M., He, L., 2007. Biomacromolecules 8,

2142–2148.Li, Y.L., Liu, M.L., Xiang, C.H., Xie, Q.J., Yao, S.Z., 2006. Thin Solid Films 497, 270–278.Ma, W.M.J., Pereira Morais, M.P., D’Hooge, F., van den Elsen, J.M.H., Cox, J.P.L., James,

T.D., Fossey, J.S., 2009. Chem. Commun., 532–534.Morris, T.A., Peterson, A.W., Tarlov, M.J., 2009. Anal. Chem. 81, 5413–5420.Nagaoka, T., Yoshino, T., 1986. Anal. Chem. 58, 1037–1042.Nakata, E., Koshi, Y., Koga, E., Katayama, Y., Hamachi, I., 2005. J. Am. Chem. Soc. 127,

13253–13261.Nicotra, V.E., Mora, M.F., Iglesias, R.A., Baruzzi, A.M., 2008. Dyes Pigments 76,

315–318.Pickup, J.C., Hussain, F., Evans, N.D., Sachedina, N., 2005. Biosens. Bioelectron. 20,

1897–1902.Russell, R.J., Pishko, M.V., 1999. Anal. Chem. 71, 3126–3132.Safavi, A., Maleki, N., Farjami, E., 2008. Biosens. Bioelectron. 24, 1655–1660.Sato, K., Anzai, J., 2006. Anal. Bioanal. Chem. 384, 1297–1301.Seneart, D.F., Teller, D.C., 1981. Biochemistry 20, 3076–3083.Sugawara, K., Shirotori, T., Hirabayashi, G., Kuramitz, H., Tanaka, S., 2004. J. Elec-

troanal. Chem. 568, 7–12.Wild, S., Roglic, G., Green, A., Sicree, R., King, H., 2004. Diab. Care 27, 1047–1053.

Xu, J.J., Zhou, D.M., Chen, H.Y., 1998. Electroanalysis 10, 713–716.Yin, X.B., Liu, D.Y., 2008. J. Chromatogr. A 1212, 130–136.Yu, C., Yam, V.W.-W., 2009. Chem. Commun., 1347–1349.Yuan, J., Wang, K., Xia, X., 2005. Adv. Funct. Mater. 15, 803–809.Zhang, X., Wang, G., Zhang, W., Wei, Y., Fang, B., 2009. Biosens. Bioelectron. 24,

3395–3398.